Development of Independent Assessment Tool at NOAA/NESDIS/STAR/JCSDA

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Monday, 5 January 2015
Deyong Xu, NOAA/NESDIS/STAR/JCSDA, College Park, MD; and K. Kumar and S. A. Boukabara

Independent Assessment Tool (IAT) we are developing here at NOAA/NESDIS/STAR/JCSDA is a utility tool by combining a few existing utility tools that analyze data assimilation results and model forecast results. Some of these existing tools are part of NOAA/NWS/NCEP operations while others are used for research purpose only. The main components of IAT include Radiance Monitoring, GRIB Extremes, Verification Database, Forecast Differences, GSI Diagnostic data analysis and Hurricane track/intensity. With these tools, research scientists can review data assimilation analysis results and verify NWP forecasts. For instance, Radiance Monitoring and GSI Diagnostic data analysis are to analyze diagnostic data from GSI analysis so radiance data assimilation performance could be monitored and potential assimilation problems could be identified. GRIB Extremes can be used to compare analysis results among different data assimilation experiments and various data assimilation systems.

However, it's not easy to set up these utilities on the various HP cluster platforms, many of these utilities require specific configurations such as job submission method, libraries used, etc., depending on the platform they are installed on. Some of the tools are not SVN-controlled, which makes it hard to track the update history of source code and to find where to get them. So what we are trying to achieve in the development of IAT is to put these utilities all together in one place being SVN-controlled, fully test them on various HP cluster platforms, generate user installation guide, make GUI to interact with one or more of the utilities to facilitate configurations, push analysis images to web server for review and generate a combined report of the most important figures created from these utilities. Using IAT will not only benefit research scientists by releasing them from dealing with software-related issues to improve the efficiency in analyzing data assimilation result and/or NWP forecast results, but also provide a comprehensive report that can best summarize the impact of their work from all aspects.